The 1380 randomly chosen New Jersey adolescents born between 1961 and 1969 are followed longitudinally at 3-year intervals. The 698 males and 682 females were tested initially in 1979-81 at the age of 12, 15, or 18. The first retest of Wave 1 (1979), Wave 2 (1980), and Wave 3 (1981) has been completed yielding a 3-year followup rate of 95%. The second retest of Wave 1 participants began April 1, 1985. A second retest of Wave 2 and Wave 3 participants and a third retest of all participants will be completed during the 5-year period (1986-91) of the proposed project. A wide range of alcohol, marijuana, and other drug use behavior (including consummatory behaviors, motivations for use, contexts of use, and consequences of use) are assessed. In addition, a broad sample of intra- and extrapersonal factors (including physiological functioning and health status, psychological states and traits, cognitive functioning, and social-cultural influences) are assessed. The selection of these variables is based on two criteria: (a) their importance in the development of alcohol and other drug using behaviors as demonstrated in previous empirical and theoretical work, and (b) their relevance in relation to various aspects of human development during adolescence and young adulthood which may be affected by substance use. With the availability of complete test-retest data for 1308 participants, research foci have shifted from cross-sectional to longitudinal analyses. Specifically, analyses of key use variables are concerned with an assessment of differences in status and change parameters of those variables in relation to (a) theoretically relevant antecedents and putative etiological factors, and (b) self-perceived and more objectively assessed consequences. Analyses of cross-sectional data and limited longitudinal analyses have produced significant results. The present proposal requests continued support for collection of additional longitudinal assessments and for ongoing analyses of the accumulating longitudinal data base. Valid findings such as those generated from this project will provide a knowledge base significant in its own right and one relevant with regard to the design and delivery of intervention programs.